From our earliest days, we are voracious and deft observers of ordinary human behavior. This consumption of overt acts is often in the service of intuiting unseen mental states. Work by Meltzoff and colleagues (e.g., Meltzoff & Brooks, 2001), for instance, shows that by even 6 months of age infants know that a grabbing hand entails wanting. When adults view an impoverished cartoon in which animated dots represent a person's arm and hand knocking on a door, they can readily intuit the phantom knocker's fear, excitement, and happiness (Pollick, Paterson, Bruderlin, & Sanford, 2001). From the smallest arch of a teasing eyebrow to the profound slouch of a disinterested onlooker, we fluidly intuit what others think from even the smallest scraps of static and dynamic evidence.
Scholars have also observed that perceivers are attuned to more elaborate and prolonged arcs of behavior as cues to underlying intentions. Social psychologists have long recognized that perceivers intuit an actor's intentions on the basis of behavior in the face of situational forces (e.g., Heider, 1958; Kelley, 1967) and from the pattern of options he or she chooses and forgoes (Jones & Davis, 1965; Newtson, 1974; see also Malle, 2004, for an overview).
Recent work has brought new attention to the role of situations in intuiting motives. Reeder and colleagues have shown that the same behaviors can imply vastly different motives depending on context (e.g.,
Reeder, Kumar, Hesson-McInnis, & Trafimow, 2002; see also Reeder & Trafimow, Chapter 7, this volume). For instance, when the same act of aggression is performed in a situation where it is provoked rather than one where it is self-serving, perceivers reach substantially different inferences about the aggressor's motives and morality. Meanwhile, Kammrath, Mendoza-Denton, and Mischel (in press) have highlighted that perceivers attend to profiles of behaviors across multiple situations to disambiguate underlying motives, showing that mental state inferences flow not just from behavioral base rates (e.g., how frequently someone is pleasant) but also from patterns across contexts (e.g., frequency of being pleasant to authorities vs. peers).
As targets of other people's perceptions, people are more than actors—they are reactors. Emotional displays are often behaviors about behaviors: a person beams when proud of her work, blushes when embarrassed by her acts, and scowls when foiled in her pursuits. In this sense, people's emotional displays are not so much intentional goal-driven behavior as unintentional reactions that stem from the match between one's goals and one's own behavior and outcomes. For the mindreader perceiving a target, these displays can reveal much, not only about passing emotional states but also about these underlying enduring motives and goals.
In recent work, my colleagues and I (Ames, Johar, & Kammrath, 2004) have explored how a target's affective displays shape inferences not only about his or her emotions but also about his or her broader intentions. Our research explored how a target's displays of self-evaluative emotions, such as embarrassment and guilt, lead perceivers to ascribe intentions to the target that were at odds with the target's overt behaviors. Extending self-presentational ideas from Goffman (1959), we argued that emotional displays help perceivers intuit how the actor regards his or her behavior and outcomes: certain negative displays can signal "that's not me" (discounting), while positive ones can signal "that's definitely me" (augmenting). Our results show that displays of self-evaluative negative affect, such as shame, after harm lead to more positive impressions of harm-doers, compared to neutral or positive affect displays, for targets ranging from violent criminals to colleagues who declined to provide a favor (Ames et al., 2004). Conversely, displays of positive affect (vs. negative affect) while helping lead to more positive impressions of helpers. Both the discounting and augmenting effects on general impressions were mediated by inferred mental states: remorseful harm-doers and cheerful helpers were both seen as having more prosocial underlying motives.
But can these affective displays endlessly correct for engaging in harmful behaviors? Certainly not. Perceivers appear to be reasonably forgiving in isolated cases, recognizing that actions and outcomes do not necessarily correspond to a target's underlying intentions. In effect, an immediate and heartfelt apology or even a pained nonverbal expression from a new acquaintance is seen as more indicative of his innocent intentions than the red wine he has just spilled on our rug. Indeed, we may even like a genuinely and appropriately remorseful spiller (diagnosis: benign) more than a harmless guest (diagnosis: uncertain). But at some point repeated harm takes over, and the stain, as it were, becomes impossible to remove. This yields a first contingency:
Affect qualifies behavior in the near term: perceived remorseful affect can lead to ascriptions of good intent to harm-doers in the short run, but repeated harm drives long-run ascriptions of bad intent.
In one study, we asked participants to imagine being paired with a classmate on a school project (Ames et al., 2004). Participants then read about a series of negative episodes involving the classmate (e.g., she showed up late for a meeting), recording their mental state inferences and impressions after each one. For some participants, the episodes were matched with photos of the classmate displaying negative self-conscious emotions, such as embarrassment and shame. For others, the photos were neutral. After the initial episode, participants seeing the negative affect displays had far more positive impressions of the target and generally assumed positive intentions (for instance, agreeing that the target wanted to be kind and respectful). After three episodes of harm (e.g., breaking the participant's cell phone, failing to complete work), impressions and mental state inferences were increasingly negative in both display conditions, though they were still significantly more positive in the remorse display condition. However, by six negative episodes (e.g., forgetting the participant's name, delivering bad work), the gap had closed: regardless of affective display, all participants had strongly negative assumptions about the target's mental states and personality. The repeated acts of harm outweighed the repeated displays of self-conscious emotions.
In sum, mindreaders often work from a target's behavior, considering not only local action (e.g., an act of help) but also arcs of behavior over time and across situations (e.g., helping authorities but not peers). Mindreaders also look to displays of emotions to intuit how actors feel about their own behavior and outcomes. In the short run, these affective displays may augment or discount behaviors; but in the longer run, behaviors appear to speak more loudly than affect.
Was this article helpful?